NEWS: Students Developing Automated Drone to Measure Soil Quality on Farms
Oliners in the lab of Kenechukwu Mbanisi, assistant professor of robotics engineering, are continuing their work in agricultural robotics by developing an automated drone that can help farmers better and more effectively measure and monitor their soil health.
Kenechukwu Mbanisi, assistant professor of robotics engineering, and his students are continuing their previous work in agricultural robotics by developing an automated drone that can help farmers better and more effectively measure and monitor their soil health.
Pictured: Members of the research team are shown on the Oval in front of the Miller Academic Center (MAC). *Not all team members were present at the time of the group photo.
Last year, Mbanisi’s students collaborated with certified-organic Powisset Farm in Dover, MA, to create a weeding robot that uses artificial intelligence (AI) to reduce the need for chemical fertilizers. This year, they are working together on a new way to help meet the distinctive needs of small- to medium-sized farms using engineering.
“Most of the work we do is stakeholder-led—we talk to the farmers, ask about their pain points, and think about how we can leverage technology to create solutions that offer value in an accessible way,” says Mbanisi. “One of the issues that surfaced was field and soil monitoring. This is a critical way that farmers get a pulse on what’s going on in the soil—moisture content, temperature, pH and nutrient levels—but it’s also time-consuming to do manually and can be challenging to prioritize on farms that can have small teams.”
The students came up with an automated drone system that flies to predetermined locations throughout the farm and collects soil data through on-board sensors. The drone then flies back to its docking station where it uploads its data to the system, which is transferred to a web-based dashboard where farmers can access the information on their phones.
“The system is designed to free up time for the farmer and deliver regular data in an easily understood format that allows them to make better-informed decisions about irrigation, fertilization, and more,” says Mbanisi.
Two students on the agricultural robotics research team, led by Mbanisi, are shown flying the team's drone on campus. The automated drone can help farmers better and more effectively measure and monitor their soil health.
Joe Leedy ’26, a mechanical engineering major, is one of the students working on the actuator that deploys the sensors into the soil.
“We are starting with just the moisture sensor for now, but the hope for the future would be to have multiple sensors on board to measure lots of different aspects of the soil,” says Leedy, who also worked with Mbanisi on last year’s weeding robot.
“It’s been really fun working on projects in agriculture because it feels like an important field to develop technology for. Making decisions on irrigation and soil additives is really important and it can be hard for farmers, so being able to help them with data for that is a huge deal,” says Leedy.
Pictured: the automated drone system created by the team.
“There are a lot of interesting problems for us to solve, such as how to incorporate a downward-facing camera to help with proper landings, or including a GPS and a barometer on board so the drone has a better sense of where it is in space,” says Dexter Friis-Hecht ’26, an electrical and computer engineering major who is working on the drone’s automated control system.
“In industry, there is a lot more interest in visual perception in robotics and drones’ ability to see their environment—it’s a cornerstone of robotics, so this kind of applied skill translates directly to possible future careers,” says Friis-Hecht.
The team is currently in the prototyping phase and hopes to do testing at Powisset next year; a longer-term goal is to be able to deploy this system operationally at the farm. Next year, Mbanisi and the team will also enter the drone system into the Farm Robotics Challenge—a competition that the lab’s weeding robot won in 2024.